grammar analysis

GE Pointers | Colin | June 3rd, 2001

Hi Suzanne,

Pinkie may have already answered some of these issues for you (what with time differences and weekend breaks and so on), but here's my view for what it's worth.

You asked for some pointers:

SFPCA is not the only analytic tool.

I find Theme/Rheme, marked/unmarked, agnation and clause movement can help determine boundaries between the parts of the sentence.

Using one of your examples:

"The Tory leader was accused of encouraging race-hate."

Try theme predication:

It was of encouraging race-hate that the Tory leader was accused.
It was race-hate that the Tory leader was accused of encouraging.
It was encouraging race-hate that the Tory leader was accused of.

All seem acceptable.

Or pronominalization:

He was accused of it.
Of it, he was accused.

*He was accused it. - suggests that "of encouraging race-hate" is a prepositional adjunct, as does the previous example.

(De-)passivization:

They accused the Tory leader of encouraging race-hate.

Marked theme:

Of encouraging race-hate, the Tory leader was accused.
Encouraging race-hate, the Tory leader was accused of [I don't think so].

Or agnation:

The Tory leader was said to be encouraging race-hate.
The Tory leader was told he was encouraging race-hate.

This suggests a Verbal or Behavioural Process of some kind (with a projected clause?)

The Tory leader was accused - of encouraging race-hate.

The Tory leader [=Receiver]

was accused [=Process: verbal]

of encouraging race-hate [=Proposal: locution]

Or is "encouraging race-hate" an Attribute being applied to the Tory leader by person or persons unknown....

Alternatively, you could just go for your SFPCA!!

The Tory leader [=Subject]

was [=Finite]

accused [=Predicator]

of encouraging rate-hate [=Adjunct]

So, what's the answer? I think the point is that there are as many answers are there are analytic tools, and Halliday introduces lots of those. Probably, the key is finding the tool that is appropriate for the kind of task you want to carry out.

Wherein lies the assignment!

I hope this doesn't muddy the waters further. Let me know if you want specific page references to "get your neck round" any of the above.

BFN

Colin

Re: Again, Pinkie (GE) | Suzanne | June 3rd, 2001

Hi Pinkie (hope you managed to get out into the sun for a while!),

1) OK, so the two highest-level clauses are:

alpha - The Tories plan to house asylum-seekers in converted holiday camps

beta - to meet their pledge to detain everyone claiming sanctuary in Britain.

Is this right?

THINK SO, YES. I HAVE THESE TWO AS THE MAIN ONES AND THREE ? EMBEDDED.

2) Do we then label the two clauses in alpha as follows? alpha alpha- The Tories plan alpha beta - to house asylum-seekers in converted holiday camps [Not too sure about this].

ARE EMBEDDED CLAUSES LABELLED FOR ALPHA BETA FUNCTIONS? I CAN'T FIND ANY EXAMPLES OF THIS IN THE UNIT. I FIND EXAMPLES OF EMBEDDED CLAUSES BEING LABELLED If THe EMBEDDED CLAUSE ITSELF IS A COMPLEX: ie UNIT 5 p 50 IN THE EXTRA EXERCISES

'Continual wars between the states made it difficult ((to farm thw fields// and maintain safe trade))'.

Where this is one clause(ie no hypotaxis or parataxis. There are several other examples, but I can't find one where an embedded clause receives this label. I would just say ALPHA and BETA.

Halliday says 'quite a lot' about embedding, but also uses alpha beta for groups and not just for clauses. ie beta for premodifier and alpha for head. He does say on p242 (1985) that:

"it is important to distinguish between embedding on the one hand and the 'tactic' relations of parataxis and hypotaxis on the other.' Basically, ' there is no direct relationship between the embedded clause and the clause within which it is embedded...'.

So I hope this means what i think it means- you can look up the rest of the paragraph.

3) Regardless, I still think that "to house asylum-seekers in converted holiday camps" is the Cdo (cf Fig 10.8, Bloor & Bloor p 211); though of course a subsidiary SFPCA analysis can be carried out within this clause. [You wrote: my original problem was how to classify 'to house' ie if this is P- it seems not? I think it's P in the subsidiary analysis, ie within alpha beta]. In other words...

alpha - [S = The Tories] [FP = plan] [Cdo = to house asylum seekers in converted holiday camps]

alpha beta - [P = to house] [Cdo = asylum seekers] [A = in converted holiday camps]

YEP- think so!!

D'you agree?- I'd certainly welcome Tom or Sue's input on this! (Or anybody else's, for that matter).

4) In addition, I guess...

beta alpha = to meet their pledge

beta beta = to detain everyone claiming sanctuary in Britain.

SEE ABOVE?, I GUESS IT HINGES; YET AGAIN ON RANKSHIFT: IF ONE STICKS TO THE PARAGRAPH I HAVE JUST QUOTED; THEN ONLY ALPHA AND BETA; AS 'to detain everyone' and 'claiming sanctuary in Britain' I WOULD HAVE AS RANKSHIFTED?

5) 'People slaved in atrocious conditions to make these pious citizens rich.' Yes, I think the TO-clause here is embedded (ie alpha-beta structure), and agree that it can't function as adjunct (my confusion!).

6) 'The Tory leader was accused of encouraging race-hate'

S = [The Tory leader], F = [was], P = [accused], A = [of encouraging race-hate]

No?

So these function as adjuncts, you're saying. Have you got a ref for me there? I can't find a word about it anywhere. This and the next are basicall the same, are they not? So, the 'on' and the 'of' here begin a Prep phrase, one as complement ('on' bogus asylum-seekers) and one as Acir? Do you agree?

7) 'to crack down on bogus asylum-seekers'

Should "on bogus asylum-seekers" be analysed as a prepositional phrase?

My opinion ... The verb "crack down" could conceivably be used intransitively, and then there's the noun "crackdown"; but you couldn't possibly say "they've cracked down on" without an object, so I'd argue that "on" should be considered as part of the prepositional phrase.

Thanks ever so much for taking the time to thrash this one out with me, Pinkie. It is much appreciated!! I think, as far as I can tell, the basic problem is the idea of rankshift- what is and waht is not - and how (and if) to label it when you've identified it. I've written to Sue Garton about this and other things, but would be pleased for input from ANYONE on the list!!

Meanwhile, have a good rest of weekend...

Suzanne- wet, windswept (park - children - the joys of motherhood!) and feeling slightly frustrated.

Re: GE, me again, sorry. | Colin | June 3rd, 2001

Hi Pinkie and Suzanne,

I cannot say I'm 100% comfortable with the analysis you two thrashed out for the Torie asylums.

How does this grab you?

Two clauses alpha and beta - agreed:

•  Alpha: 'The Tories plan to house asylum-seekers in converted holiday camps'
•  Beta: 'to meet their pledge to detain everyone claiming sanctuary in Britain.'

Alpha: Main clause with hypotactic projection (p250).
'We are going to house asylum-seekers in converted holiday camps,' said the Tories.

The Tories plan - to house asylum-seekers in converted holiday camps.

The Tories {=S} plan {=F/P} to house asylum-seekers {=C} in holiday camps {=A}?

[proposal : idea : resolving] p291.

Beta: dependent clause hypotactically enhancing alpha (p237) containing an embedded projection and an embedded modifier.

Why are they doing it?

'to meet their pledge' - enhancement because it is [cause:purpose] (iv) p237

to meet {=P} pledge {=C}?

'their pledge to detain everyone' is a nominal group with pledge as the head and 'to detain everyone' as the postmodifier.

their pledge {head} to detain {P} everyone {C}

[proposal : locution : promising] p291.

'everyone claiming sanctuary in Britain' is a nominal group with everyone as the head and 'claiming sanctuary in Britain' as the postmodifier.

everyone {head} claiming {P} sanctuary {C} in Britain {A}.

So, as I see it, two separable clauses which are related hypotactically, alpha and beta.

Alpha is a clause complex with two hypotactically related clauses, alpha beta is a projection of alpha alpha.

Beta is a clause complex which is a hypotactic enhancement of alpha. Beta contains a complement which has two embeddings: an embedded projection which modifies the complement, and an embedded clause which modifies the complement of the embedded projection.

Short of drawing up boxy tables and so on, this is the best I can come up with in terms of a straightforward analysis.

Phew! Bit complex that one!

BFN

Colin

Again, Pinkie (replying to Suzanne) | Pinkie | June 3rd, 2001

Suzanne...

I'm not at all sure about this: I need help from Tom or Sue!

First off, we have three mutually exclusive categories, no?

1) Paratactic clause - main clause linked to another eg by AND

2) Hypotactic clause - dependent (subordinate) clause

3) Embedded clause - forms part of another clause Yes?

I analyzed "The Tories plan to house asylum seekers in converted holiday camps" as a single clause with embedded Complement: [S = The Tories] [FP = plan] [Cdo = to house asylum seekers in converted holiday camps] [Whether or not this is the correct analysis, you're right that alpha and beta labelling don't seem to be used for embedded clauses].

However, B&B p 206 seems to suggest that sentences of this type should be analysed as main + dependent, ie the TO-clause is a hypotactic projection:

alpha - [S = The Tories] [FP = plan] beta - [P = to house] [Cdo = asylum seekers] [A = in converted holiday camps].

Is the analysis with the TO-clause as embedded Complement not permissible in the Hallidayan system? Intuitively, I like it better as embedded than as hypotactic: it seems to me an integral and "contained" part of the sentence, not an optional add-on like "to make these pious citizens rich".

[By the way 1- another my-mistake: I said that the TO-clause in "People slaved in atrocious conditions to make these pious citizens rich" was embedded, but I was clearly wrong, it's hypotactic. Sorry!].

[By the way 2 - 'of encouraging race-hate' as Adjunct, cf B&B p52 (S = Loewi) (F = was) (P = obsessed) (Acir = by the idea)].

Hope you're enjoying this as much as I am. Though I agree, combining grammar with parenthood is not entirely straightforward!

Pinkie

Re: Colin (GE) | Suzanne | June 3rd, 2001

Thanks for the quote, Colin. I know where you're coming from and I will abandon things when it gets too sticky, don't worry. Nor do I intend (for the other person who mentioned theme and rheme, to stick to SFPCA- in fact, my assignment will probably be more Theme/Rheme and Process/Participant based. However, (there had to be one), it is difficult to go into Theme/Rheme in detail without having qorked out the way in which the sentences fir together - that's the basic rationale. Don't worry, though, I won't risk my sanity in any way whatsoever. Haven't no got any time to reply to the other terrific comments, but will try and do so soon....

Suzanne

Hypotaxis vs. Rankshifty | Colin | June 3rd, 2001

If anyone has the patience to keep up with this... I wonder whether the distinction between hypotactic clause and rank-shifted clause may be something that is not widely accepted outside Hallidayan circles. Consider the following constructed sentence:

He wanted to buy a new suit, to wear to the wedding.

Halliday I think requires us to analyse this as follows:

alpha (main): He wanted to buy a new suit

beta (hypotactic expansion, ie dependent): to wear to the wedding

Within alpha - alpha alpha (main): He {S} wanted {FP} alpha beta (hypotactic projection): to buy {P} a new suit {Cdo} within beta - to wear {P} to the wedding {A}.

Is this right?

But e.g. Quirk et al and Huddleston* I think would analyze this as follows:

Main: He wanted to buy a new suit

Dependent: to wear to the wedding

Within main - He {S} wanted {FP} to buy a new suit {Cdo} to wear to the wedding {A} (i.e. "to buy a new suit" is embedded).

This latter approach seems to me much more intuitive and natural: it allows the terminal non-finite clause to act as adjunct (cf replacement with "for the wedding"); and it treats "to buy a new suit" as a verbal complement embedded within the main clause (cf replacement with "a new suit").

In more formal terms, the Hallidayan system I think can be summarized as follows:

1) Paratactic

2) Hypotactic: a) expansion, b) projection

3) Embedded (= rank-shifted).

The system implicit in Quirk et al and Huddleston appears to be:

1) Paratactic

2) Hypotactic: a) expansion (dependent clause), b) embedded (inc projection) This seems to me to make much more sense.

In this connection, note the following from Butler "Systemic Linguistics: Theory and Applications" (1985) p96, about SFG linguist Fawcett: "Fawcett [...] allows the A element [...] to be filled by a clause, thus dispensing with the distinction between alpha beta and rank shifted structures maintained by Halliday".

In other words, the original analysis I gave for Suzanne's sentence ** wasn't a correct Hallidayan analysis, but I think Halliday's approach here can be and has been challenged. Though I'm open to suggestions as to the advantages of distinguishing between hypotaxis and rank-shift!

Pinkie

•  Huddleston R 1994, Introduction to the Grammar of English, CUP.

** [The Tories = S] [plan = FP] [to house asylum-seekers in converted holiday camps = Cdo] [to meet their pledge to detain everyone claiming sanctuary in Britain. = A, of purpose].

Oh GOD! (GE) | Colin | June 5th, 2001

Hello Pinkie, Colin and anyone else who's following this...

Colin- The analysis that you have given of the sentence looks very convincing. You quote p250 for the Alpha: main clause with hypotactic projection. The Tories plan - to house asylum-seekers in converted holiday camps.

But it is not an independent clause, is it. And I come back to a basic question - is projection a different level of analysis? Can I not 'just' ask for a clause-complex analysis? Can an embedded clause be given a hypotactic label? Halliday p. 242, bottom paragraph (please read it yourself, it's quite long) seems to me to say that you cannot. On p. 250 (your quote) I can't find an embedded clause with a hypotactic label, either.

I would go along with the rest, and I can't really see where it differs from our analysis significantly, i.e. Beta to meet their pledge to detain everyone claiming sanctuary in Britain.

'to meet their pledge' dependent clause (hypotactic) P/C

'to detain everyone' embedded clause functioning as postmodifier of nominal group with pledge as head(P/C)

'claiming sanctuary in Britain' embedded clause as postmodifier with everyone as head.

Apart from the thing with the embedding, I think we basically agree. As I said in my last entry, the problem seems to be rankshift- and I'm really NOT arguing with matters of projection, either. But CAN this EMBEDDED clause have a hypotactic function? I still have not found an answer or an example of this(in MAKH). If they can't, then we are still in agreement, except that 'to house asylum-seekers' will be an embedded clause functioning as C within the clause, no?

(I wish I hadn't started this!)

Suzanne

GE: Sorry, Suzanne | Pinkie | June 5th, 2001

I feel rather responsible for your desperation, since in my earlier messages I was getting hypotactic and embedded mixed up, and probably just confused things. But I'm more or less confident of my last message (GE: hypotaxis vs rankshift):

Very briefly, I agree that the correct Hallidayan analysis of the TO clause in "The Tories plan to house asylum seekers in converted holiday camps" is as a hypotactic projection (i.e. NOT embedded, NOT forming part of the SFPCA structure of the main clause, and thus NOT considered as Cdo). Hypotactic and embedded (aka rankshifted) are as far as I can see MUTUALLY EXCLUSIVE categories in the Hallidayan system. So to consider this clause as an embedded Cdo would be a perfectly reasonable but different (and non-Hallidayan) analysis; see my previous message.

Hope I'm not making things worse! Colin, what do you think about this? Rob? (And Sue and Tom of course, if you're reading this).

Pinkie

Hippo taxis and so on | Colin | June 5th, 2001

Suzanne, etc....

As I see it, briefly:

Two ways to join clauses of the same rank: Clause-complexing or embedding.

Clause-complexing can be parataxis or hypotaxis. In hypotaxis, one of the clauses is dependent on the other through some kind of projection or expansion. The expansion may be an elaboration of the main clause or an enhancement of one of the participants that has already been identified in the main clause.

We saw the boy, who ran away.

[The boy is already identified, so "who ran away" is enhancement because it provides extra information about an identified participant.]

Who did we see? The boy.
What did we do? He ran away.

Embedded clauses are down-shifted or rank-shifted and can appear in a clause or a group. The only groups they can appear in are nominal or adverbial. Embedded clauses which are modifying certain head nouns do not always need to keep the head. Embedded clauses are (always?) used to identify the participant(s).

We saw the boy who ran away.

[Here "who ran away" identifies the boy we saw, so it is embedded as a postmodifier in a nominal group]

Who did we see? The boy who ran away.
What did he do? We do not know exactly, maybe he broke a window....

Punctuation, if used correctly, often identifies the differences. The spoken language uses tone units to differentiate between the two.

As I see it, the major problem is when embedded clauses are in the clause itself. Here, they can look like complexing rather than embedding. Or vice versa.

How would you classify "that you do not understand this" in the following?

That you do not understand this is clear.
It is clear to me that you do not understand this.
The fact that you do not understand this is clear.

I would go for embedded with missing head, hypotactic projection of my thoughts, embedded (modifying "The fact") But I could be wrong!

I think it may be useful to maintain the 3-level distinction because of the role embedded clauses play in identifying participants rather than just elaborating on them or enhancing the main clause. However difficult that makes life for we grammarians - or should that be us grammarians?

Enjoy!!

Colin

GE Tory concentration camps | Tom Bloor | June 6th, 2001

Most excellent grammarians,

I can't keep up with you people. First, let me congratulate you on the high level of grammatical argumentation. Pinkie has asked for input from Sue G or me, but you seem to be managing very well without it so I'm pleased neither of us responded sooner. I won't comment on individual contributions, but I think they are really impressive. My analysis of the sinister news item is as follows:

The Tories plan to house asylum-seekers in converted holiday camps]] Alpha

to meet their pledges [[to detain everyone [[claiming sanctuary in Britain]] ]] Betaie the second clause from "to meet ....in Britain" is dependent on the first (hypotactic expansion) and, incidentally, it is a purpose clause (as at least one person pointed out). Everything after "pledges" is embedded (rankshifted non-finite clause postmodifying "pledges"); a nominalised projection; "claiming sanctuary in Britain" is embedded within the larger embedding, postmodifying "everyone"; this last is what some grammars would call a reduced relative clause (akin to "who is claiming asylum in Britain".

So far, this is in line with what some of you whiz-kids have suggested. The trickiest bit is dealing with "plan" and "to house .....". I analyze "plan to house" as a verb group complex. (cf "the Tories want to house ..."; "the Tories intend to house" cf Bloor & Bloor p29 "She wants to understand".).

So SFPCA analysis of the first clause is:

The Tories S

plan to house F/P^P (ie a fused Finite/Predicatorfollowed by a Predicator)

asylum-seekers Cdo in converted holiday camps Acir.

The issue of verbal group complexes is sadly under-addressed in B&B. Blame the publishers for limiting us on size. Unfortunately, also the index entry for "verb complex" is wrong; blame me for careless proofreading. See Halliday 1994 S7A.4 - S7A.6 pp 278-291. This account is pretty heavy and a bit confusing because he uses the labels Alpha and Beta throughout for relations inside the group complex and then in 7A.6 moves into projected clauses. His intention is to show the link between verbal group complexes and clause complexes and he presents some as alternative treatments of the same data. (As one of you recently observed, Halliday also uses Alpha, Beta notation for relations other than clause complex, so when you see the Greek alphabet, don't assume that it is necessarily a clause label).

To summarize his comments very crudely: clause complex and verb complex merge into each other and/or are different ways of looking at the same data. See also Martin, Matthiessen & Painter 1997 pp 116-117; 177-178. See also the brief discussion in GE Unit 05 pp34-35. If it seems that I have been keeping this important concept quiet, I apologize. It was not intentional.

If as some of you assumed, "to house asylum seekers ..." were to be analyzed as a dependent clause, it would indeed be Alpha Beta with "The Tories plan" as Alpha Alpha. My problem there would be in deciding what kind of dependent clause it is.

Verbal group complex also solves Pinkie's problem with "He wanted to buy a new suit".

He wanted to buy a new suit Alpha

He S

wanted to buy F/P^P (VG complex)

a new suit || Cdo

to wear to the wedding Beta

to wear P

to the wedding. Acir.

Yes, in most other grammars, e.g. Quirk, transformational grammar, almost any grammar, the distinction between embedded and dependent clauses does not exist. This is pointed out in GE Unit 05. Even in some SFG grammars (Fawcett, Downing) some of Halliday's dependent clauses are treated as embedded (see Bloor & Bloor p 192-193). It is not just an arbitrary distinction, though. There are grammatical reasons for making it, but as usual with grammar, there are borderline cases that present problems. I must admit that I found it one of the most difficult things to accept in SFG. Certainly, it would make analysis easier if every dependent clause was treated as embedded, but I am not convinced it would make for a better analysis.

Maybe you have continued this debate since I downloaded my mail. If so, I shall try not to come back. Rest assured. I don't intend to join in all your GE debates. You are doing very well without staff participation, and, after all, I have retired.

By the way, we are currently working on a new edition of The Functional Analysis of English, to be completed as soon as we have Halliday's proofs for his new edition of IFG. If you have any suggestions, please let me know. I have identified a number of errors, but I don't mind hearing about them again or any I haven't noted. Obviously more on verbal complexes is called for, but we are strictly forbidden to increase the length of the book (to keep the price down) - every author's nightmare - so suggestions for things to take out or shorten would also be welcome. I'm not saying we'll be willing or able take your advice, but we might. No acknowledgments, I'm afraid; we've reached the limit there.

Keep analyzing!

(Independent clause: Verbal group complex: imperative)

Tom Bloor

Re: GE Thanks Tom | Pinkie | June 6th, 2001

A thousand thanks to Tom for directing us to verbal group complexes, of which I for one was blissfully ignorant. Will read! By the way, Tom has recently corrected another of my efforts, to whit: in a previous post I suggested that "Nearby is Chandor Beach" can be analysed as S-FP-Cint, but in fact it should be Cint-F-S (as supported by subject-verb agreement: Nearby is Chandor Beach, Nearby are several beaches). That it's F not FP is because Halliday treats finite BE as Finite only (see Tom's OTHER message).

Pinkie

Spain

Re: me again (GE) | Suzanne | June 10th, 2001

Hello everyone!!! (I won't address this to anyone in particular, but there are a few parties out there more 'interested' than others).

First, thanks to Tom and Sue for their input. The verbal group complex tip is invaluable. My analysis is going OK and I have moved on to projection/expansion in the text, something which I didn't seem to have grasped at first reading. On one of my test sentences (02.06 I think), Colin sent me a reply analysis of the sentence 'The Tory leader was accused of encouraging race hate' - I haven't got the e-mail anymore, or I'd include the answer. Colin very rightly pointed out that SFPCA is not the only analytic tool(to which I answered that I wasn't doing an assignment on SFPCA, but that I felt I needed to have an overview of the texts for all the functions - even if superficially). Colin then went on to suggest theme/rheme - prediction. He finally came up with an SFPCA analysis of 'of encouraging race-hate' as an adjunct by comparing it with other possible positions and ways of expressing the same meaning. However, having read the first paragraph of unit 5 (p19)- Tom clearly states that 'Halliday does not allow for a clause to function as an adjunct'. This would seem to contradict that analysis unless you do not see it as a clause.

If we look at this as a non-finite dependent clause, might it be an expansion?

Hague was accused (alpha)

of encouraging race-hate.(beta projection)

(I am not at all convinced, please feel free to tear strips off me!).

Or what about a clause complex?

Hague was(F) accused of (P) encouraging(P) race-hate ©.

Would that be so different from the 'plan to house' example?- (Of course, this assumes that 'of' is part of the verb).

Hmmm...

Suzanne

Dependent/embedded clauses & adjuncts | Tom Bloor | June 11th, 2001

To clarify (I hope):

The point is that a DEPENDENT clause cannot be an Adjunct (or a Subject or Complement) in the clause on which it is depends because it is a separate clause. However, an EMBEDDED clause can be an S or C or part of S or C or part of an A of the clause in which it is embedded because it is part of that clause. So traditional adverb clauses (Time: when clause: Reason: because clause; concession: although clause, etc "When he arrived" are treated by Halliday as dependent expansion clauses and not as embedded clauses realising Adjunct.

The suggestion made that "of encouraging race-hatred" is an Adjunct is not incompatible with that view since the embedded clause is only "encouraging race-hatred" not the whole of the prepositional phrase. The implied view here is that the Adjunct is realized as a prepositional phrase: "by [[encouraging race-hatred]]" with the nonfinite embedded clause as 'object of the preposition'.

Hence the suggestion is:

The Tory leader is accused of encouraging race-hatred.

The Tory leader: Subject

is : Finite

accused: Predicator

of encouraging race-hatred: Adjunct (cf "of racism".

Note: "encouraging race-hatred": embedded clause in a PP).

Transitivity analysis:

"The Tory leader": Target

is accused: Verbal Process.

"of encouraging race-hatred": Circumstance.

NB. the Sayer is unspecified.

Your alternative analysis of a dependent projection clause is the other obvious contender, but at the moment I feel less comfortable with that than the one above.

Your Verbal Group complex analysis (I assume that is what you meant to say below) is less appealing though not completely daft, by any means. You have:

>Hague (S) was(F) accused of (P) encouraging(P) race-hate ©.

Given Halliday's somewhat arbitrary cut-off point between clause complex and verbal group complex even structures like "want her to go" or "expect her to answer" are treated as two clauses by Halliday (though he agrees it is debatable). But "accuse" has no single participant equivalent to "I want to go" or "I expect to answer".

I don't feel happy with "accuse of" as a lexical verb. I think you would have been less likely to go for this analysis if it had been active, say:

Opponents accused Hague of encouraging race-hate.

I would be slightly happier (but not entirely happy, for some of the above reasons) with:

Hague (S) was (F) accused (P) of encouraging race-hate (A).

So the first analysis is the one I would opt for.

Best

Tom Bloor

Prepositions/particles, embedding/dependency | Tom Bloor | June 11th, 2001

Before I withdraw from these discussions, here is a quick response to Suzanne's much earlier mailing (below).

I don't see any way of treating "to crack down on bogus asylum-seekers" that is preferable to "crack down" as a phrasal verb and "on bogus asylum seekers" as a prepositional phrase. I agree that "on" could be perceived as identifiable with "crack down" in some strong collocational way at least, but by far the most straightforward way of dealing with this is the above method.

In fact, the whole question of preposition versus verb particle is very confused and probably insoluble. Often instances are very clear-cut but often they are not. A couple of months ago, there was prolonged discussion of similar data on a specialist discussion list I subscribe to, and the various experts came up with very different views on how to label the items concerned. So, Suzanne, you shouldn't worry if you can see more than one way of analyzing it. There are lots of fuzzy edges.

To confirm the answer to another question that has arisen in the great clause complex discussion: yes, embedded clauses are the same clauses as rankshifted clauses (I think this was probably resolved by you and it is stated in the file). The two words don't say quite the same thing but they refer to the same thing. I think "embedded" comments on the syntactic structure (it's tucked into something else) and "rankshifted" comments on the status (it's a clause but it's being used as/in something of a lower rank). Don't worry about it. Treat them as the same. It is the dependent clause that is different, being attached to the dominant clause and not part of it.

Re hypotaxis: the term "hypotactic" is applied to dependent clauses but not usually applied to embedded clauses. I don't see any principled reason for this, but it helps to prevent further confusion. Therefore, Beta, Gamma, etc labels are not applied to embedded clauses - unless, of course, they contain an embedded clause complex: e.g.

She offered the explanation [[that she had been working hard ||to prepare for the exam.]]

Here, WITHIN THE EMBEDDING, "that she had been working hard" is Alpha and "to prepare for the exam" is Beta.

Also: "People slaved in atrocious conditions to make these pious citizens rich". You rightly analyzed "to make these pious citizens rich", and one of you pointed out that, since this was a dependent clause of purpose (not an embedded clause), it could not be Adjunct in the other clause; the two clauses have separate SFPCA structure. This is perfectly correct (though the mistake of thinking it is an Adjunct is very understandable; in grammars which do not distinguish between embedded and dependent clauses, it could well be analyzed as an Adjunct).

At least, with the Conservative Party so totally defeated in the recent election, we don't have to worry about their introducing concentration camps for asylum-seekers though we might have to watch that the new government doesn't adopt the idea. Not to mention those thousands of fascist voters in Oldham.

Tom Bloor

 

Archive Categories